Modeling study suggests diabetes screening cost-effective starting between 30 and 45

It is cost-effective to screen asymptomatic patients for type 2 diabetes beginning between the ages of 30 and 45 and repeating the test every three to five years, according to a computer model.

The study used data from a representative sample of the U.S. population to create a simulated population of 325,000 people 30 years of age without diabetes. Diabetes incidence, health care costs associated with screening and treatment, and potential complications (including myocardial infarction, stroke and microvascular complications) were simulated for the following 50 years of the hypothetical patients' lives. A number of different screening strategies were put into the model—screens starting at different ages, occurring at different frequencies, or triggered by a diagnosis of hypertension.

Compared to not screening, all of the strategies reduced the incidence of myocardial infarction and microvascular complications and increased the number of quality-adjusted life-years (QALYs). There was little to no effect on stroke incidence. The screening strategies varied more in their cost per QALY. Screening at frequent intervals (every six months starting at 30 or every year starting at 45) or late-onset screening (every three years starting at 60) was associated with a cost of more than $15,000 per QALY.

Screening patients who were already being treated for hypertension had the lowest cost per QALY, but less effect on microvascular outcomes because fewer cases of diabetes were diagnosed. The most cost-effective strategies appeared to be those that started screening patients between 30 and 45 and repeated screening every three to five years, the researchers concluded. Such strategies were associated with a cost per QALY of $10,500 or less. The study was published online by The Lancet on March 30.

The cost of these screenings could be minimized further by conducting the screening in conjunction with other recommended care, such as lipid screening, the authors noted. Although the model is limited by being a simulation, the study's results are applicable to real life, the study authors said, concluding that screening every three to five years starting between 30 and 45 would be cost-effective for the U.S. population.

The study did not address the American Diabetes Association's current recommendations, which suggest screening 30- to 45-year-olds only if they have risk factors, an accompanying comment noted. (The study authors did mention that a risk assessment before screening could improve cost-effectiveness.) The results of the simulation do provide further evidence that diabetes screening should be combined with screening for hypertension and lipids, the comment concluded.